Boosting partial least‐squares discriminant analysis with application to near infrared spectroscopic tea variety discrimination

Potential of NIR spectroscopy for predicting cellulose nanofibril quality in commercial bleached Kraft pulp of Eucalyptus

Near infrared (NIR) spectroscopy with chemometrics was used for rapid source identification and component quantification of Arnebiae Radix. The modeling parameters were determined using the processing trajectory method. The quantitative model of target components was established using the model fusion method. The root mean square error of calibration (RMSEC), root mean square error of validation (RMSEP), determination coefficient of calibration (Rcal 2 ) and validation (Rpre 2 ), and the ratio of standard error of prediction to standard deviation (RPD) were used to evaluate the constructed model. The RMSEC, Rcal 2 , RMSEP, Rpre 2 , and RPD of β-β'-dimethylacryloacanin partial least square (PLS) model established using the step-by-step strategy were 1.3004 mg/g, 0.9453, 1.3833 mg/g, 0.9544, and 4.6650, respectively. The RMSEC, Rcal 2, RMSEP, Rpre 2 , and RPD of total pigment PLS model were 4.5095 mg/g, 0.9533, 4.5679 mg/g, 0.9534, and 3.0509, respectively. The RMSEC, Rcal 2 , RMSEP, Rpre 2 , and RPD of β-β'-dimethylacryloacanin Bagging-PLS model established using the process trajectory were 1.6126 mg/g, 0.9159, 1.2254 mg/g, 0.9649, and 5.2662, respectively. The RMSEC, Rcal 2 , RMSEP, Rpre 2 , and RPD of the total pigment Bagging-PLS model were 4.3499 mg/g, 0.9565, 4.4662 mg/g, 0.9116, and 3.1205, respectively. The accuracy of quantitative model may be improved using the process trajectory and model fusion method. Partial least squares discriminant analysis (PLS-DA) was used to establish the discriminant models of different sources of Arnebiae Radix. The predicted accuracy rate was 100%. The method is simple, rapid, and demonstrated for the discrimination and quantitative analysis of the main active ingredients of Arnebiae Radix.

Moving window partial least-squares (MWPLS) regression was coupled with near infrared (NIR) spectra as an interval selection method to improve the performance of partial least squares discriminant analysis (PLSDA) models. This method was applied to the identification of artificial bezoar, natural bezoar and artificial bezoar in natural bezoar and compared with some traditional pattern recognition methods, such as principal component analysis (PCA), linear discriminant analysis (LDA) and PLSDA. The introduction of MWPLS enhanced the performance of PLSDA model. The results obtained showed that moving window partial least-squares discriminant analysis (MWPLSDA) can extract wavelength intervals with useful information and build simple yet effective classification models that can significantly improve the classification accuracy. Then MWPLSDA was used to identify natural bezoar by geographical origin; a promising result was achieved. The work showed that MWPLSDA could be a promising method for quality analysis and discrimination of chinese medical herbs according to geographical origin.

Discrimination of Camellia japonica cultivars and chemometric models: An interlaboratory study

This study evaluates the feasibility of near infrared (NIR) spectroscopy to distinguish between different cartilage injury types associated with post-traumatic osteoarthritis and idiopathic osteoarthritis (OA) induced by mechanical and enzymatic damages. Bovine osteochondral samples (n = 72) were subjected to mechanical (n = 24) and enzymatic (n = 36) damage; NIR spectral measurements were acquired from each sample before and after damage, and from a separate control group (n = 12). Biomechanical measurements were then conducted to determine the functional integrity of the samples. NIR spectral variations resulting from different damage types were investigated and the samples classified using partial least squares discriminant analysis (PLS-DA). Partial least squares regression (PLSR) was then employed to investigate the relationship between the NIR spectra and biomechanical properties of the samples. Results of the study demonstrate that substantial spectral changes occur in the region of 1700–2200 nm due to tissue damages, while differences between enzymatically and mechanically induced damages can be observed mainly in the region of 1780–1810 nm. We conclude that NIR spectroscopy, combined with multivariate analysis, is capable of discriminating between cartilage injuries that mimic idiopathic OA and traumatic injuries based on specific spectral features. This information could be useful in determining the optimal treatment strategy during cartilage repair in arthroscopy.

Screening analysis to detect adulteration in diesel/biodiesel blends using near infrared spectrometry and multivariate classification

Analytical capabilities of chemiresistive microsensor arrays in a simulated Martian atmosphere

This paper establishes a novel and rapid method for detecting pure melamine in milk powder using near infrared (NIR) spectroscopy based on least squares-support vector machine (LS-SVM). Partial least square discriminant analysis (PLS-DA) was used for the extraction of principal components (PCs). The scores of the first two PCs have been applied as inputs to LS-SVM. Compared to PLS-DA, the performance of LS-SVM was better, with higher classification accuracy, both 100% for the training and testing set. The detection limit was lower than 1 ppm. Based on the results, it was concluded that NIR spectroscopy combined with LS-SVM could be used as a rapid and accurate method for detecting pure melamine in milk powder.

Pre-harvest screening on-vine of spinach quality and safety using NIRS technology.

Species of the Amaranthus genus are very versatile and have potential for the application in the development of commercial products. The near infrared spectroscopy (NIR) is an efficient tool that can help in the quality control of products, quickly and non-destructive to the sample. The goal of this study was to carry out the distinction of seed and oils of different Amaranthus species using the near infrared spectroscopy. Three species were used: A. viridis L., A. hybridus L. e Amaranthus sp. (commercial). The spectra acquired from the sample using the near infrared spectroscopy were submitted to the partial least squares discriminant analysis (PLS-DA) and to the principal component analysis (PCA). Through PCA, it was possible to differentiate the Amaranthus species both for seeds and oils. Through PLS-DA, it was possible to predict the classes of the species with high degree of correct classification, with 96.67% of correct classifications for seeds and 98.89% for oil. Thus, with the use of the near infrared spectroscopy associated with the multivariate statistical analysis, it is possible to classify the different Amaranthus species, especially when using the oil.

The problems of adulteration and mislabeling are very common in the food industry. Laser-induced breakdown spectroscopy (LIBS) coupled with chemometric methods has many intrinsic advantages on adulteration analysis of various materials. In this work, several chemometric algorithms, i.e., principal component analysis (PCA), decision tree (DT), random forest (RF), partial least squares discriminant analysis (PLS-DA), linear discriminant analysis (LDA), and support vector machine (SVM), were carried out assisted by LIBS technology to study the classification performances on rice geographic origins. A series of samples, including 20 kinds of rice samples from different geographic origins, was detected using LIBS with no pretreatment processes. For data analysis, PCA was employed to reduce the input variables, and to reduce the collinearity of LIBS spectral results as well. The results showed the classification accuracies of the mentioned chemometric algorithms of DT, RF, PLS-DA, LDA, and SVM with 89 input variables of 86.80%, 96.30%, 96.80%, 98.60%, and 99.20%, respectively. At the same time, the operation times of these algorithms were 3.81 s, 54.64 s, 3.63 s, 2.09 s, and 531.01 s, respectively. On the other hand, 30 principal components of input variables were also tested under the same conditions. The classification accuracies for the above algorithms were 81.60%, 98.00%, 95.70%, 98.40%, and 99.20%, respectively. The operation times were 2.01 s, 4.88 s, 3.67 s, 0.36 s, and 308.55 s, respectively. In addition, the five-fold cross-validation classification accuracies with 30 input variables for DT, RF, PLS-DA, LDA, and SVM were 83.75%, 97.95%, 94.75%, 98.35%, and 99.25%, respectively. As a result, LDA was demonstrated to be the best and most efficient tool for rice geographic origin classification assisted by LIBS with high accuracy and analytical speed, which has great potential for rapid identification of adulterated products in agriculture without use of any chemical reagent.

The aim of this paper was to evaluate the effect of two different matrices (e.g., starch base flour vs. protein base flour) on the ability of near-infrared (NIR) spectroscopy to classify binary mixtures of chickpea (protein), corn and tapioca (starch) flours. Binary mixtures were made by mixing different proportions of chickpea plus corn, chickpea plus tapioca, and corn plus tapioca flour. Spectra were collected using NIR spectroscopy and the data analyzed using techniques such as principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA). The results showed an effect of the matrix on the PLS-DA classification results, in both classification rates and PLS loadings. The different combinations of flours/mixtures showed changes in absorbance values around 4752 cm−1 that are associated with starch and protein. Nevertheless, the use of NIR spectroscopic might provide a valuable initial screening and identification of the potential contamination of flours along the supply and value chains, enabling more costly methods to be used more productively on suspect samples.

Yerba mate samples from three different states of Brazil were evaluated in order to discriminate them regarding the presence of sugar and geographic origin. High-performance liquid chromatography (HPLC), phytochemical compounds, in vitro antioxidant activity, visible and near-infrared (NIR) spectroscopy, colorimetry, and electronic nose were used in tandem with chemometric methods. The multiblock exploratory analysis (ComDim) was able to discriminate the samples containing sugar; however, it was not possible to discriminate them by geographical origin. Furthermore, ComDim results showed the NIR spectra presented the best discriminating capacity. Partial least square discriminant analysis (PLS-DA) models constructed using NIR spectra classified the samples assertively according to the presence of sugar (100% of sensitivity and specificity for the prediction set), and reasonable models were also obtained for the geographic classification (80% of sensitivity and 93% of specificity for the prediction set). The multiblock approach allowed an overall evaluation of the data collected through different analytical methods. In addition, among the methods applied, NIR spectroscopy was faster and cheaper and allowed for better sample discrimination.

Classification of geographical origins and prediction of δ13C and δ15N values of lamb meat by near infrared reflectance spectroscopy

Near-infrared (NIR) spectroscopy was used to distinguish between game meat from six different species, i.e. three medium-sized (impala, blesbok and springbok) and three large-sized (eland, black wildebeest and zebra) that were harvested (collected and slaughtered) from different farms across South Africa. Longissimus thoracis et lumborum (LTL) muscle steaks were removed and scanned with a handheld NIR spectrophotometer in the spectral range of 908 to 1700 nm. Spectra were treated with two different pre-processing combinations: (1) smoothing, standard normal variate and de-trending (SNV-Detrend), and (2) SNV-Detrend and Savitzky-Golay 2nd derivative. Data were explored with principal component analysis (PCA) and classified with linear discriminant analysis (LDA), soft independent modelling by class analogy (SIMCA) and partial least squares discriminant analysis (PLS-DA). For discrimination and classification, models were developed within each of the medium- and large-sized groups. LDA resulted in good classification accuracies ranging from 68 to 100%, irrespective of the pre-processing combination used. PLS-DA performed well (classification accuracies ranging from 70 to 96%) when spectra were treated with SNV-Detrend and Savitzky-Golay 2nd derivative. The prediction results obtained with SIMCA, pre-processed with smoothing and SNV-Detrend, ranged from 67% (springbok) to 100% (impala and eland). Although accurate models were obtained, they could still be improved by extending the sample set with meat samples from each species to cover variation in terms of season, geographical location, age and sex.